Reflashing one of my Linksys nodes with 1.1.2. the load failed all I get now is:

(1) rapid flashing power led

(2) slow blinking DMZ led

(3) no connection can be made to the router

Is there a recover method I can use to reflash and put it back in the network?

73 de mike

My recommendation would be to try a tftp install.  The bottom of the firmware installation instructions explains how to do it. 

Basically as you first apply power to the node you have a around a one second window to execute a tftp command to install new firmware.  I have used this method many times to recover from a bad install or a lost password. I do not recommend a 30-30-30 reset. It is almost guaranteed to brick your system.

Clint, AE5CA

thanks for the help. but  (seems to always be a but)...

i don't know the IP address of the node. does it default to something at power on?

73 de mike

If you have flashed the node but not set it up, then http:// 172.27.0.1:8080 should get you in to the node do the setup. 

You might want to refer to:
http://www.broadband-hamnet.org/images/stories/hsmmmesh-step-by-step.pdf

and

http://www.broadband-hamnet.org/documentation/68-firmware-installation-instructions.html

Clint, AE5CA

When the router firsts boots up, the CFE tells it to have a static IP of 192.168.1.1 and a netmask of 255.255.255.0 (regardless of the firmware). If the boot_wait flag is enabled (located in the CFE), you can TFTP firmware to the router at this address during the 1 second window at boot. Make sure you set a static IP address on your computer too.

If the previous method does not work, let me know your make and model of your router, I will tell you the pins we need to short and we'll have to go through it that way.

If you have a JTAG cable, let me know too.

[KF5JIM 2014-08-14- 04:24:40]:
> you can TFTP firmware to the router at this address during
> the 1 second window at boot. Make sure you set a static IP
> address on your computer too.

Seems like this could be a more consistent way to deal with routers in some arbitrary firmware state.

Questions:
When does this one second window start? Upon DC, before/after one of the LED's, or ?
What user action has to happen within the one second window to get "in"?

Can the "standard" BBHN or stock BIN file be TFTP'd over via the bootstrap or does it have to be a special one?

Guessing that if something is corrupted to the point the window does not happen, what mode does the CFE go into when you "short" something?  I didn't see initialization detection/failure defined in a cursory review of the Broadcom CFE spec.

Mark K5LXP
Albuquerque, NM

And I would agree. That would be the more consistent way to deal with routers in some states, but from my experience, there have only been a few routers that have the boot_wait flag enabled by default. From my understanding, when you load the mesh firmware, it enables this flag on first boot after flashing, but I have not had much luck with it.

The 1-second window starts as soon as you apply DC power.

What user action has to happen within the one second window to get "in"?
Read this: http://www.dd-wrt.com/wiki/index.php/Tftp_flash

You can TFTP whatever firmware you want. It's up to the router to read, accept and/or reject what you send.

When you short two pins on the flash chip, it erases the chip (minus the CFE) and puts the router into a TFTP recieve state until readable firmware is sent to the router. If the CFE is corrupt, your only method is to erase the flash chip and JTAG the correct CFE back to the router...at which time (after finishing the CFE write) the router goes into TFTP recieve mode.

No joy... just spent two hours trying to catch that one second window.

i can ping 192.168.1.1

i can setup tftp as defined as a .sh and it executes, but no workie.

this is a Ver 354G

i have no jtag cable but could order one. they are on amazon, just need to know which one and then what to do with it.

I can try the pin short process.

73 de mike

Short pins 15 and 16 on the flash chip. Please take a look at this post for more details: http://www.wi-fiplanet.com/tutorials/article.php/10724_3618616_3/Bricked-Or-How-to-Resurrect-a-Dead-Linksys-WRT54G.htm

After you do this, set your computer to a static IP and TFTP the firmware back to the router at the 192.168.1.1 address.

As for JTAG, look here:
http://www.broadband-hamnet.org/hsmm-mesh-forums/view-postlist/forum-1-general/topic-1095-can-this-be-used-for-jtag.html#ccbp4870

still no joy...

my netbook drops the connection with power off and takes too long on power up to make the one second window.

I think this one is a destine for the recycle bin. I appreciate all the help. The good new ... my other 7 nodes are working fine.

73 d mike

An Ethernet switch might help here. Plug your computer into the switch. Then connect the node to the switch. 

The switch will keep your computer port alive when you power cycle the switch. 

The instuctions Paul sent you has a script for doing the tftp from a Linux computer. That can make it easier as well. Start the script and keep power cycling the node until it takes. 

Sorry you have had this much trouble getting this node up and running. 

Clint, AE5CA

Greetings from Iceland!

You really should not be having this much trouble. The fact that your Ethernet port drops connection when you turn off the router tells me that you very well might still have DHCP enabled on your computer. Please check and make sure that you have set a static IP address of 192.168.1.x (where x can be anything in the appropriate IP allocation range except for 1) and a subnet mask of 255.255.255.0.

Here is what I would do...Unplug your Linksys device, plug in the Ethernet cable between your computer and the device, set a static IP address in the same subnet, get everything setup with the TFTP programme, apply power and start the upload.

finally got back to trying to fix this.

setup =

(1) bricked 54g

(2) switch

(3) Linux net book

using TFPT script as defined on site.

process =

setup net  book as 192.168.1.5/255.255.255.0/192.168.1.1

connect cat5 cables

start script point to bbhn-1.1-2brcm-2.4-squashfs.trx

power up 54g .... ERROR MESSAGE

received ACK <block=0>

sent data <block=1, 512 bytes>

received error <code=4, msg=code pattern incorrect>

error code 4: code pattern incorrect

script stops...

What do I need to fix this one? I am getting closer, I think...

73 de mike

I believe your tftp command is trying to upload the wrong file.  Make sure you match the file you use to your router.  I would think that you will need the WRt54G factory version.  The .trx version is for when you have bbhn on the node previously.

Clint, AE5CA

BINGO!!!!

Much joy, thanks for the help.

73 de mike